1. Field of the Invention
The present invention relates to a job execution managing method and job execution managing system comprising a storage device which has a plurality of physical volumes, and which is used with the respective physical volumes respectively assigned to a plurality of logical volumes, and an information processing device which instructs the abovementioned storage device to execute jobs.
2. Description of the Related Art
In the past, an efficient job re-execution technique has been proposed which uses a load dispersion function in cases where jobs are re-executed because of computer trouble in a computer system constructed from a plurality of computers. In this job re-execution technique, when a job that was in progress in the computer in which trouble occurred is re-executed, efficient execution and an early resolution of the effects of computer trouble are realized by judging the computer that is used to re-execute the job that is the object of re-execution from load information for the computers that are candidates for performing re-execution and information relating to the conditions of execution up to the time that the job that is the object of re-execution was interrupted (for example, see Japanese Patent Application Laid-Open No. 11-353284).
In a storage device, there may be cases in which the plurality of physical volumes installed in the storage device are assigned to a plurality of logical volumes and used. In such cases, if trouble occurs in one of the physical volumes, there may be cases in which the clients of the client server system using the storage device cannot recognize that trouble has occurred in this physical volume, and can only recognize that trouble has occurred in the logical volume containing this physical volume.
In such cases, when the execution of a job using the logical volume containing the physical volume in which trouble has occurred fails, the client can recognize that the execution of this job has failed; accordingly, the client can stop the execution of this job. However, in regard to the execution of other jobs in a state awaiting execution using another logical volume containing the physical volume in which the abovementioned trouble has occurred, even though this execution fails as a result, the client cannot recognize the trouble involved in this other logical volume; accordingly, there is no method for stopping the execution of these other jobs that are in a state awaiting execution. Consequently, the following problem arises: namely, jobs that will fail are needlessly executed.